1887. ] Permanganice Acvd. 111 
The difference between these numbers is considerable and 
can hardly be attributed to errors of experiment, but rather to 
the fact that the distillate obtained in the experiments was not 
constant in composition. 
The percentage of available oxygen in permanganic acid, 
H,Mn,O,, would be 33°33, or more than three times as great 
as the mean amount found by experiment. Hence the substance 
could not be pure permanganic acid. 
It is however possible that the distillate is a mixture of per- 
manganic acid with a quantity of water insufficient to liquefy it, 
aqueous vapour having been removed by the current of air from 
the diluted sulphuric acid in the retort. On this supposition the 
mixture would contain about 30 °/, of permanganic acid. 
As the product obtained by distillation of permanganate with 
slightly diluted sulphuric acid could not be regarded as a body 
of constant composition, a determination of the density of this 
product in the gaseous state appeared likely to be of little 
value. 
Acting upon the suggestion above made that the distillate 
was a mixture of permanganic acid and water, it was resolved 
to try the effect of pure sulphuric acid in the place of the mono- 
hydrated acid used in the preceding experiments. Accordingly a 
few drops of Nordhausen acid were poured upon a small quantity 
of potassium-permanganate in a test tube surrounded by ice, but 
the reaction was of such a violent nature that no furtber experi- 
ments were attempted with this reagent. 
In the next experiment pure re-distilled sulphuric acid was 
poured on to potassium-permanganate contained in a retort resting 
- in a bath of cold water. The tubulus was closed by a cork, with 
a glass tube arranged as in the experiment with the diluted acid, 
so that dry air might be passed through the liquid in the retort if 
necessary. A piece of caoutchouc tubing with a screw-clip allowed 
the tube to be opened or closed at will. The neck of the retort 
passed air-tight through a caoutchouc cork nearly to the bottom of 
a strong wide test tube. This cork was perforated for the air-tight 
passage of another tube, which allowed the gases produced in the 
reaction to escape into the atmosphere. When the pure sulphuric 
acid had been poured upon the permanganate in the retort, no 
pink vapours were given off, notwithstanding that the temperature 
was cautiously raised to 70°, and that bubbles of gas were being 
given off from the surface of the mixture. The exit tube of the 
apparatus was therefore connected up with a Bunsen air-pump 
giving a vacuum of about 450 mm. of mercury, so as to reduce the 
pressure inside the apparatus. The rate of evolution of the gases 
was at once increased, but the retort and test tube attached to it 
shewed no signs of pink vapour or distillate, notwithstanding that 
